Prevention trials demonstrated that selenium is a promising chemopreventive agent for prostate cancer. Seleniuminhibited human prostate cancer cell growth, blocked cell cycle progression at multiple transition points, and induced apoptotic cell death. We have demonstrated a novel mechanism of selenium anticancer action in which selenium markedly reduces androgen receptor (AR) expression and AR-mediated gene expression including prostate-specific antigen (PSA) in human prostate cancer cells in vitro and in vivo. Further studies demonstrated that selenium-mediated AR reduction involved in multiple processes including a reduction of AR mRNA transcription, a decrease in AR mRNA stability, and an increase in AR protein degradation. Androgen signaling through androgen receptor (AR) plays an important role not only in maintaining the function of the prostate, but also in promoting the development of androgen-independent prostate cancer. A common treatment for prostate cancer is androgen deprivation. Although most men respond to androgen deprivation therapy initially, almost all relapse due to the growth of androgen-independent cancer cells. Most of the androgen deprivation treatments are either blocking androgen-AR binding or reducing the levels of androgen. Based on our novel finding that selenium disrupts AR signaling by reducing AR expression, a completely different mechanism from current androgen deprivation therapy, it is conceivable that targeting AR signaling by a combination of androgen-deprivation therapy and selenium (reducing AR expression) might improve the efficacy of current androgen deprivation therapy. This concept was validated in vitro in which the combination of selenium and anti-androgen (Casodex) synergistically inhibited clonogenic ability of human prostate cancer cells, providing a rationale for in vivo validation of the combination of selenium and anti-androgen therapy for prostate cancer. Based on these findings, the hypothesis is that anticancer effects of selenium are mediated, in part, by inhibition of AR activity and that decreased AR signaling may reduce the incident of prostate cancer and prevent or delay relapses after androgen deprivation therapy. The goal of this application is to elucidate the importance of down regulation of AR signaling by selenium in prostate cancer chemoprevention and therapy.' To test this hypothesis, 4 aims are proposed. 1. To determine the molecular basis of the effects of selenium o'rf'AR transcriptional regulation. 2. To determine the posttranslational regulation of AR expression and activation by selenium. 3. To determine the significance of modulation of AR by selenium in prostate cancer. 4. To evaluate the combination of selenium and anti-androgen therapies in mouse models of prostate carcinogenesis.